面向删失数据的稳健治疗规则学习
Learning Robust Treatment Rules for Censored Data
Yifan Cui · Junyi Liu · Tao Shen · Zhengling Qi · Xi Chen
中文摘要
直接通过最大化期望结果来估计最优治疗规则的研究快速增长。在生物医学与运营应用中,常观察到删失生存结局,此时截断平均生存时间与生存概率备受关注。针对删失生存结局,提出两种稳健准则学习最优治疗规则:前者以给定分位数(如中位数)为截断点,最大化截断平均生存时间;后者最大化缓冲生存概率,其中预设阈值根据截断平均生存时间进行调整。开发基于采样的差凸算法学习所提最优治疗规则,并给出理论依据。模拟研究中,估计量相比现有方法表现更优。利用 AIDS 临床试验数据演示所提方法。
关键要点
- 01针对删失生存结局,提出两种稳健准则学习最优治疗规则
- 02第一种准则最大化截断平均生存时间,截断点由给定分位数指定
- 03第二种准则最大化缓冲生存概率,阈值根据截断平均生存时间调整
- 04开发基于采样的差凸算法,并提供理论依据
- 05模拟研究显示估计量优于现有方法,并用 AIDS 临床试验数据验证
解读
尚无解读。
原始英文摘要
arXiv:2408.09155v3 Announce Type: replace-cross Abstract: There is a fast-growing literature on estimating optimal treatment rules directly by maximizing the expected outcome. In biomedical studies and operations applications, censored survival outcomes are frequently observed, in which case the truncated mean survival time and survival probability are of great interest. In this paper, we propose two robust criteria for learning optimal treatment rules with censored survival outcomes; the former one targets an optimal treatment rule maximizing the truncated mean survival time, where the cutoff is specified by a given quantile such as the median; the latter one targets an optimal treatment rule maximizing buffered survival probabilities, where the predetermined threshold is adjusted to account for the truncated mean survival time. We develop a sampling-based difference-of-convex algorithm for learning the proposed optimal treatment rules, and provide theoretical justifications for them. In simulation studies, our estimators show improved performance compared to existing methods. We also demonstrate the proposed method using AIDS clinical trial data.